Dynamic calibration and adaptive signal decoding method based on PSI5 interface

Abstract

The invention relates to the technical field of signal decoding, in particular to a PSI5 interface-based dynamic calibration and adaptive signal decoding method, which comprises the following steps of: acquiring pre-calendar data and analog signals output by a sensor under different working environment conditions; performing feature extraction on the analog signal to obtain a feature parameter; preprocessing the historical signal data; dynamically calibrating the characteristic parameters of the analog signal by using a calibration rule to obtain an accurate analog signal; generating a decoding rule of the current signal by using a decision tree algorithm according to the accurate analog signal and a PSI5 protocol standard; decoding the accurate analog signal according to a decoding rule to obtain effective data; and comparing the effective data with expected data to judge whether deviation exists or not, and if yes, adjusting the calibration rule and decision tree algorithm parameters according to the deviation condition. According to the invention, by dynamically calibrating the deviation of the signal and adjusting the decoding parameter, invalid data transmission is reduced, and the signal precision is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of signal decoding, and in particular to a dynamic calibration and adaptive signal decoding method based on a PSI5 interface. Background Art

[0002] PSI5 is a bidirectional digital communication protocol widely used in the automotive electronics field. It is mainly used for data transmission between sensors and electronic control units. It uses a two-wire current interface and Manchester encoding to achieve efficient and reliable signal transmission.

[0003] Traditional signal calibration methods are often static and based on fixed parameter settings. Traditional decoding rules are based on fixed parameters and cannot adapt to signal changes under different operating conditions. The inability to flexibly adjust calibration parameters in the face of various real-time changing operating conditions during vehicle operation can easily lead to large deviations between the calibrated data and the actual signal. Conventional signal decoding methods are also mostly designed for standard ideal signals and lack the ability to adapt to signal changes in complex real-world environments. Once the signal exhibits waveform distortion due to interference or changes in signal characteristics due to sensor aging, decoding errors or even the inability to decode occur. This urgently requires a signal decoding method that is both dynamically calibrated and adaptive to overcome these shortcomings and improve the accuracy and reliability of signal processing based on the PSI5 interface. Summary of the Invention

[0004] The present invention provides a dynamic calibration and adaptive signal decoding method based on the PSI5 interface, which is used to solve the defects of low accuracy, poor robustness and resource waste caused by static calibration errors, fixed decoding rules and lack of feedback optimization in the prior art.

[0005] The present invention provides a dynamic calibration and adaptive signal decoding method based on a PSI5 interface, comprising: S1: Collect historical signal data and analog signals output by sensors under different working environment conditions.

[0006] S2: Extract features from the analog signal to obtain feature parameters. Preprocess the historical signal data to obtain predicted data.

[0007] S3: Use calibration rules to dynamically calibrate the characteristic parameters to obtain accurate analog signals.

[0008] S4: Based on the precise analog signal and the PSI5 protocol standard, a decision tree algorithm is used to generate decoding rules that are adapted to the current signal.

[0009] S5: Decode the precise analog signal according to the decoding rules to obtain valid data.

[0010] S6: Compare the valid data with the expected data, adjust the calibration rules and decision tree algorithm parameters according to the deviation, and obtain accurate and valid data.

[0011] According to a dynamic calibration and adaptive signal decoding method based on the PSI5 interface provided by the present invention, in step S1, the analog signal is converted into a digital signal through the PSI5 interface, and the digital signal includes multiple data frames, each data frame including a synchronization bit, a data bit and a check bit.

[0012] According to the present invention, a dynamic calibration and adaptive signal decoding method based on a PSI5 interface includes, in step S2, characteristic parameters including the initial amplitude, frequency, phase, and data frame format of the analog signal. Preprocessing includes outlier processing, one-hot encoding, time encoding, data normalization, and Gaussian noise addition to the historical signal data.

[0013] According to a dynamic calibration and adaptive signal decoding method based on a PSI5 interface provided by the present invention, in step S3, dynamically calibrating characteristic parameters using calibration rules includes: S31: Set the structural parameters of the timing integration model and collect real-time environment parameters.

[0014] S32: Use pre-history data to train the time series integration model.

[0015] S33: Input the real-time environmental parameters into the trained time series integration model to predict the current calibration parameters, convert the current calibration parameters into an operational calibration rule, and dynamically adjust the signal characteristic parameters according to the calibration rule.

[0016] According to a dynamic calibration and adaptive signal decoding method based on a PSI5 interface provided by the present invention, in step S32, training the time series integration model includes: S321: Divide the predicted data into training set 1 and test set 1.

[0017] S322: Setting the hyperparameters of the current time series integration model according to the structural characteristics of the time series integration model and the characteristic parameters of the predicted data.

[0018] S323: Use the embedder of the temporal ensemble model to transform the training set into a low-dimensional learnable space.

[0019] S324: In the low-dimensional learnable space, the discriminator updates the hyperparameters of the current time series integration model according to the accuracy of determining whether the received data is from the training set 1 or the fake data generated by the generator, using the gradient descent algorithm until the comprehensive evaluation index of the time series integration model decreases for two consecutive rounds, and then stops updating the hyperparameters.

[0020] S325: Use a test set to test the time series integration model until the trained time series integration model meets the preset threshold, and use the generator to generate new data and merge it with the original predicted data to obtain new predicted data.

[0021] According to a dynamic calibration and adaptive signal decoding method based on the PSI5 interface provided by the present invention, in step S324, training the timing integration model includes: during the hyperparameter update process, the generator continuously generates false data in an attempt to deceive the discriminator, and the discriminator finds the false data output by the generator, and the two repeatedly perform adversarial training. After each round of adversarial training, the timing integration model is tested using a test set, and the model performance indicators and changes in hyperparameters are monitored until the timing integration model stops updating the hyperparameters.

[0022] According to a dynamic calibration and adaptive signal decoding method based on a PSI5 interface provided by the present invention, in step S4, the specific steps of using a decision tree algorithm to generate a decoding rule adapted to the current signal are as follows: S41: Setting parameters of the decision tree algorithm according to the number of feature parameters.

[0023] S42: Construct the analog signal characteristic parameters and the correct decoding under the corresponding preset decoding rules into a training set 2 and a test set 2, and input the training set 2 into the decision tree algorithm.

[0024] S43: Splitting the best feature into nodes and constructing branches to form a decision tree. The categories corresponding to the leaf nodes of the decision tree are preliminary decoding rules.

[0025] S44: Using a set of current analog signals with known correct decoding results, the current analog signals are decoded according to preliminary decoding rules to obtain current analog signal decoding rules.

[0026] According to a dynamic calibration and adaptive signal decoding method based on a PSI5 interface provided by the present invention, in step S5, decoding the precise analog signal includes: S51: Find the starting frame in the precise analog signal according to the definition of the signal frame structure in the decoding rule.

[0027] S52: From the starting frame position, check whether the precise analog signal frame after the starting frame is complete according to the decoding rule. If so, keep it unchanged; otherwise, repair the precise analog signal frame using a signal repair algorithm.

[0028] S53: extracting analog signal data corresponding to the data segment from the complete accurate analog signal frame according to the position and length range of the data segment in the signal frame in the decoding rule.

[0029] S54: Convert the analog signal data segment into digital signal data according to the decoding conversion method for the data segment in the decoding rule.

[0030] S55: Analyze and restore the digital signal data according to the data format in the decoding rules to obtain valid data.

[0031] According to a dynamic calibration and adaptive signal decoding method based on a PSI5 interface provided by the present invention, in step S53, the data segment includes: Position-related data: used to indicate the specific position of the data segment in the entire signal frame.

[0032] Length range data: the length range occupied by the data segment itself.

[0033] Coding characteristic data: the coding characteristics of the data segment in the form of analog signals.

[0034] Decoding conversion data: Describes the specific conversion method of how the data segment is converted from analog signal to digital signal.

[0035] Data format data: The data format requirements followed when the data segment is finally parsed and restored.

[0036] According to a dynamic calibration and adaptive signal decoding method based on the PSI5 interface provided by the present invention, in step S6, after adjustment, the parsed parameters are error checked to determine whether the error check result exceeds the threshold. If so, redundant data is used to compensate for the valid data.

[0037] The present invention provides a dynamic calibration and adaptive signal decoding method based on the PSI5 interface. It uses calibration rules to dynamically calibrate the characteristic parameters of analog signals and decode accurate analog signals. This method solves the low accuracy defects caused by static calibration errors and fixed decoding rules in the prior art. The beneficial effects achieved are as follows: This invention uses dynamic calibration to monitor sensor output signal deviations in real time and adjusts decoding parameters through algorithms to ensure signal accuracy. Traditional fixed calibration methods struggle to cope with complex and changing real-world operating conditions. This method, however, uses differential signal acquisition to acquire the original analog signal and applies dynamic calibration rules to correct characteristic parameters in real time. Differential signal acquisition effectively suppresses common-mode interference and ensures the integrity of the original signal. Dynamic calibration adjusts calibration parameters based on real-time operating conditions, ensuring consistently high analog signal accuracy.

[0038] The PSI5 interface in the present invention adopts Manchester encoding and current modulation transmission, and has inherent anti-interference capabilities. Adaptive decoding is further optimized through algorithms to dynamically adjust the decoding threshold or filtering parameters to adapt to noise interference under different working conditions. Adaptive decoding can identify the signal characteristics of different sensors and achieve "plug and play". There is no need to configure decoding parameters separately for each sensor, reducing system complexity. The adaptive decoding rule generation mechanism based on the PSI5 protocol can dynamically adjust the decoding strategy according to the current signal characteristics and channel status. As a standard interface for automotive sensor networks, the PSI5 protocol, with its flexible frame structure and encoding method, provides the basis for adaptive decoding. The system automatically selects the optimal decoding algorithm by analyzing parameters such as the frequency, duty cycle, and noise level of the signal, combined with a preset decoding rule library.

[0039] The adaptive decoding in the present invention can dynamically adjust the decoding frequency or sampling rate according to the signal strength, avoid unnecessary computing overhead, and extend the life of the vehicle battery. Dynamic calibration can compress signal redundancy, reduce invalid data transmission, and further improve bus utilization in combination with the efficient encoding method of the PSI5 interface. By comparing the decoded data with the expected value in real time, the system can quickly identify deviations and trigger closed-loop adjustments. The parameter correction of the dynamic calibration rule is based on the results of the deviation analysis, such as adjusting the gain coefficient or compensation factor to offset environmental interference. The optimization of the decision tree algorithm parameters iteratively updates the model weights through machine learning technology, so that the decoding rules continue to approach the optimal state. If the verification fails, the retransmission mechanism or redundant data compensation strategy can ensure data integrity and avoid system failure due to a single decoding error.

[0040] The dynamic calibration process in this invention monitors sensor health in real time, identifies potential faults in advance, and sends warnings via the vehicle network. The automated nature of adaptive decoding reduces the need for manual intervention and lowers maintenance costs, making it particularly suitable for large-scale fleet management. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 This is a step diagram of a dynamic calibration and adaptive signal decoding method based on a PSI5 interface provided by an embodiment of the present invention; Figure 2 This is a flowchart of using predicted data to train a time series integration model provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0044] The following combination Figure 1-Figure 2 The present invention describes a dynamic calibration and adaptive signal decoding method based on the PSI5 interface.

[0045] like Figure 1 As shown, an embodiment of the present invention provides a dynamic calibration and adaptive signal decoding method based on a PSI5 interface, comprising: S1: Collect historical signal data under different working environment conditions, use differential signals to collect the analog signal output by the sensor, and convert it into a digital signal stream through the PSI5 interface. The analog signal contains multiple data frames, each of which contains synchronization bits, data bits, and check bits.

[0046] The synchronization bit is a key component of a data frame, used to synchronize clocks and data between the transmitter and receiver. During signal transmission, the transmitter sends data according to a specific clock rhythm, and the receiver must also accurately receive and interpret the data according to the same rhythm. The synchronization bit helps the receiver "find the right rhythm." It helps the receiver determine the starting position of a data frame, enabling subsequent data bits to be correctly identified and processed, avoiding data misalignment and misjudgment, and ensuring accurate and orderly data transmission.

[0047] Data bits are the core component of a data frame that carries the actual information being transmitted. They contain the various useful data that the sender wants to convey to the receiver. This data can represent different meanings depending on the specific application scenario. For example, in a sensor network, data bits may contain numerical encodings of physical quantities such as temperature, humidity, and pressure collected by the sensor. In a communication system, data bits may be the encoded binary representation of information such as text, images, or audio. After receiving the data bits, the receiver uses corresponding decoding rules to restore them to the original information content to ensure effective information transmission.

[0048] The parity bit is used to detect errors during data transmission. Since signals may be subject to various interferences during transmission, such as electromagnetic interference and noise, which can cause certain bits within the data to change, the parity bit helps the receiver determine whether the received data is correct. Using a specific parity algorithm, the sender generates a parity bit based on the data bits and adds it to the data frame. The receiver then uses the same algorithm to perform a checksum calculation on the received data, comparing the calculated result with the received parity bit to determine if there are any transmission errors, thereby improving data transmission reliability.

[0049] S2: Extract features from the collected initial signal to obtain characteristic parameters such as the signal's initial amplitude, frequency, phase, and data frame format.

[0050] Amplitude is a key metric for describing signal strength or size. For different types of signals, such as analog electrical signals and mechanical vibration signals, amplitude represents the magnitude of the signal in the corresponding physical dimension.

[0051] Frequency refers to the number of times a signal repeats its periodic changes per unit time, measured in Hertz. It is a key parameter that characterizes the periodic characteristics of a signal. Signals of different frequencies often have different physical sources and functions. In the field of communications, electromagnetic waves of different frequencies are used to transmit different types of information. In mechanical systems, the vibration frequency of equipment components can reflect their operating status and the presence of potential faults such as looseness or wear. Normally functioning equipment often has a vibration frequency within a specific, reasonable range.

[0052] Phase describes the position of a signal relative to a reference point at a given moment, often expressed as an angle. Phase plays a key role when comparing or synthesizing multiple signals.

[0053] The data frame format primarily relates to the organization of digital signals during transmission and storage. It specifies how signal data is segmented, identified, and what additional information is included. Different application scenarios often require different data frame formats. For example, in communication networks, the data frame format determines how data is sent, received, and parsed, ensuring accurate information transfer between devices. When sensor networks collect data and transmit it to a host computer, the data frame format specifies how the data collected by each sensor is organized, enabling the host computer to accurately identify and process data from different sensors.

[0054] S3: Using the calibration rules and the real-time monitored environmental parameters, the characteristic parameters of the signal are dynamically calibrated to obtain accurate signal characteristic parameters.

[0055] like Figure 2 As shown, dynamic calibration includes: S31: Set the structural parameters of the timing integration model and collect real-time environment parameters.

[0056] S32: Use pre-history data to train the time series integration model.

[0057] S321: Divide the predicted data into a training set 1 and a test set 1 in a ratio of 8:2. Other ratios may also be used.

[0058] S322: Set the model's hyperparameters based on the structural characteristics of the time series integration model and the characteristic parameters of the predicted data. These hyperparameters include the hidden layer dimension, the number of attention heads, and the learning rate. The hidden layer dimension determines the model's ability to abstractly represent features. The number of attention heads affects the model's focus on different features and time steps. The learning rate determines the step size for parameter updates during model training. An appropriate learning rate helps the model converge faster and more stably.

[0059] S323: Use the embedder of the temporal ensemble model to transform the training set into a low-dimensional learnable space.

[0060] S324: In the low-dimensional learnable space, the discriminator updates the hyperparameters of the current time series integration model according to the gradient descent algorithm based on the accuracy of determining whether the received data is from the training set or fake data generated by the generator. The hyperparameter update stops after the comprehensive evaluation index of the time series integration model decreases for two consecutive rounds. During the hyperparameter update process, the generator continuously generates fake data in an attempt to deceive the discriminator, and the discriminator identifies the fake data output by the generator. The two repeatedly conduct adversarial training. After each round of adversarial training, the time series integration model is tested using the test set. The model performance indicators and changes in hyperparameters are monitored until the time series integration model stops updating the hyperparameters.

[0061] S325: Test the time series ensemble model using the test set to determine whether the trained time series ensemble model meets the preset threshold. If so, the generator of the time series ensemble model generates new data and merges it with the original data to obtain new predicted data. Otherwise, adversarial training continues. To prevent overfitting of the ensemble model, the new data set is divided using a 5-fold cross-validation method.

[0062] S33: Input the real-time environmental parameters into the trained time series integration model to predict the current calibration parameters, convert the current calibration parameters into an operational calibration rule, and dynamically adjust the signal characteristic parameters according to the calibration rule.

[0063] S4: Based on the calibrated signal characteristic parameters and in combination with the PSI5 protocol standard, a decision tree algorithm is used to generate decoding rules adapted to the current signal.

[0064] S41: Set the evaluation criteria for node splitting of the decision tree algorithm and parameters such as the maximum depth of the tree to provide a reasonable initial framework for the operation of the algorithm, so that it can effectively carry out the learning and adjustment process towards generating adaptive decoding rules.

[0065] S42: Construct the analog signal characteristic parameters and the correct decoding under the corresponding preset decoding rules into a training set 2 and a test set 2, and input the training set 2 into the decision tree algorithm.

[0066] S43: The decision tree algorithm is started to learn and reason, generating preliminary decoding rules based on the input data. Based on the input training data set, the algorithm continuously selects the best features for node splitting and constructs branching paths, gradually forming a decision tree. The categories corresponding to each leaf node are preliminary decoding rules.

[0067] S44: preparing a set of independent analog signal test samples with known correct decoding results, and performing decoding operations on these test samples according to the generated draft decoding rule to obtain the decoding rule.

[0068] S5: Decode the calibrated signal according to the decoding rules to obtain valid data output by the sensor.

[0069] The specific steps for decoding the precise analog signal according to the decoding rules are: S51: Based on the definition of the signal frame structure in the decoding rules, look for the iconic features in the analog signal that represent the start of the frame, such as a pulse of a specific amplitude or frequency as a frame header mark. Use signal analysis software or algorithms to accurately lock the frame header position, and then determine the range of the entire signal frame, preparing for the subsequent extraction of the area where valid data is located.

[0070] S52: Following the determined frame header, check whether the subsequent signal parts are complete and meet expectations according to the frame length, order of each part, and other requirements specified by the decoding rules. If any missing or abnormal parts are found, promptly analyze whether it is a problem in the acquisition and transmission link or an abnormality in the signal itself, and try to solve it by re-acquisition or using signal repair algorithms to ensure the availability of the entire signal frame.

[0071] S53: According to the position and length range of the data segment in the signal frame clearly specified by the decoding rules, accurately extract the analog signal part corresponding to the data segment from the complete signal frame. For example, if the rules indicate that the data segment starts from the 8th sampling point after the frame header and is 20 sampling points long, then strictly follow this standard to intercept it.

[0072] Position-related data: This data clearly indicates the specific location of the data segment within the entire signal frame. For example, in a simple analog signal frame structure, the signal frame includes a frame header, a data segment, and a frame trailer. The position data of the data segment will indicate the sampling point from which it starts and ends. This precise position definition facilitates subsequent steps to accurately lock in the target range to be extracted from the complete analog signal frame. For example, in a signal frame with a length of 100 sampling points, specifying that the data segment is located between the 20th and 80th sampling points is the embodiment of its position-related data.

[0073] Length range data: This defines the length range of the data segment itself. This, combined with location information, further accurately identifies the data segment. For example, the length of a data segment can be fixed at 60 sampling points, or it can be variable within a certain range, such as a minimum length of 50 sampling points and a maximum length of 70 sampling points. Length range data not only facilitates accurate extraction, but also allows for a certain degree of flexibility in data segments under different circumstances, ensuring that the corresponding data segment can be correctly obtained in a variety of scenarios that meet the rules.

[0074] Coding characteristic data: This describes the coding characteristics of the data segment in analog signal form, specifically how it carries and represents information at the analog level. For example, a specific modulation scheme is used to encode the original information to form this data segment, or different values are represented according to a specific pattern of level changes. This coding characteristic data is an important basis for subsequent decoding and conversion. Only by clearly understanding the coding characteristics in its analog state can it be accurately converted to its corresponding digital form.

[0075] Decoding and conversion data: This includes data on the specific conversion method for converting the analog signal into a digital signal for this data segment. For example, this data specifies the use of a specific quantization algorithm to convert the analog signal's amplitude into a digital value in quantization units of 0.1V. Alternatively, it specifies a mapping relationship, such as where a certain range of analog signal levels corresponds to a specific code in the digital signal. These explicit conversion rule-related data guide the analog-to-digital conversion operation in step S54.

[0076] Data format-related data: This relates to the data format requirements followed when the data segments are finally parsed and restored. For example, it may require that the converted digital signal data segments be arranged in a specific byte order, or that they be further parsed according to a specific protocol format to ensure that valid data that meets expectations can be ultimately restored for subsequent use and processing.

[0077] S54: Based on the specific decoding conversion method for the data segment in the decoding rules, the extracted analog signal data segment is converted into a corresponding digital form or other form that meets the subsequent analysis requirements. For example, if the data segment is encoded using pulse width modulation, it is converted into a corresponding digital code based on the correspondence between the pulse width and the specific numerical value, completing the key data conversion step.

[0078] S55: After verification and error processing, the signal data is parsed and restored in detail according to the decoding rules that specify the data format and the meaning of each part. For example, the rules stipulate that the first few digits represent the sensor's temperature measurement value, and the middle few digits represent the humidity measurement value. Based on this correspondence, the data is restored to specific measurement values, generating the valid data content output by the sensor. The parsed valid data is checked for rationality, such as whether the value is within the reasonable physical range. If any unreasonable data is found, further investigation is carried out to determine whether it is caused by a problem in the decoding process or an abnormality in the sensor itself, and appropriate treatment is taken if necessary.

[0079] S6: Compare the valid data obtained by decoding with the expected data. If there is a deviation, adjust the calibration rules in the dynamic calibration step and the decision tree algorithm parameters in the adaptive decoding rule generation step according to the deviation feedback, and perform error check on the parsed parameters. If the check fails, trigger the retransmission mechanism or enable redundant data compensation.

[0080] Compare the valid data and expected data one-to-one along the corresponding dimensions. For example, in a time series, the difference between the valid temperature data collected at the same time and the pre-set expected temperature data at that time can be calculated, or the trend of change can be observed to determine whether there is a discrepancy between the two. Common comparison methods include calculating statistical indicators such as difference, relative error, and mean square error. For example, the difference between the valid temperature data and the expected temperature data at each time point can be calculated. If the absolute value of the difference is within a certain allowable range, the data can be considered to be basically consistent. If the difference exceeds this range, it indicates a deviation.

[0081] Once a deviation is determined, the cause of the deviation is analyzed. Causes include zero-point drift of the measuring device, the influence of environmental factors, and aging of internal system components. For example, in a weighing system, if the measured weight data is always found to be larger than the actual expected weight, after investigation, it is found that the sensor has experienced zero-point drift due to long-term use, resulting in a deviation in the initial measurement benchmark. Based on the cause of the deviation, the calibration rules are modified accordingly. Taking the weighing system as an example, if the problem is zero-point drift, the zero point of the sensor needs to be recalibrated, that is, the relevant parameters for zero-point correction in the calibration rules need to be modified to ensure that subsequent measurements are performed with the correct benchmark so that the valid data can be closer to the expected data.

[0082] Decision tree algorithms can dynamically adjust their parameters based on changes in input data to adapt to new situations and better achieve desired goals. For example, adaptive learning rate algorithms in machine learning adjust the learning rate based on the error during each training session, enabling faster model convergence and better results.

[0083] When a deviation between the valid data and the expected data is detected, information such as the magnitude and direction of the deviation is fed into the decision tree algorithm as feedback. For example, in an intelligent flow control system, if the valid data deviates significantly from the expected flow rate and the actual flow rate is too low, the decision tree algorithm will appropriately increase the relevant parameters controlling the valve opening based on the degree of deviation to adjust the flow rate in the expected direction. By continuously updating the algorithm parameters based on the deviation feedback, the valid data is ultimately brought into close agreement with the expected data.

[0084] The present invention provides a dynamic calibration and adaptive signal decoding method based on the PSI5 interface. It uses calibration rules to dynamically calibrate the characteristic parameters of analog signals to decode precise analog signals. This method solves the defects of the prior art caused by static calibration errors, fixed decoding rules, and lack of feedback optimization, such as low accuracy, poor robustness, and waste of resources. The beneficial effects achieved are: Dynamic calibration monitors sensor output signal deviations in real time and uses algorithms to adjust decoding parameters to ensure signal accuracy. Traditional fixed calibration methods struggle to cope with complex and changing real-world operating conditions. This method, however, uses differential signal acquisition to acquire the original analog signal and applies dynamic calibration rules to modify characteristic parameters in real time. Differential signal acquisition effectively suppresses common-mode interference, ensuring the integrity of the original signal. Dynamic calibration adjusts calibration parameters based on real-time operating conditions, ensuring consistently high analog signal accuracy.

[0085] The adaptive decoding rule generation mechanism based on the PSI5 protocol dynamically adjusts the decoding strategy based on current signal characteristics and channel conditions. As a standard interface for automotive sensor networks, the PSI5 protocol's flexible frame structure and encoding provide the foundation for adaptive decoding. The system automatically selects the optimal decoding algorithm by analyzing signal parameters such as frequency, duty cycle, and noise level, combined with a pre-defined decoding rule library.

[0086] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units. That is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0087] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods of each embodiment or certain portions of the embodiments.

[0088] Finally, it should be noted that the above embodiments are intended only to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will appreciate that modifications may be made to the technical solutions described in the aforementioned embodiments, or that some of the technical features may be replaced with equivalents. Such modifications or replacements do not deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A dynamic calibration and adaptive signal decoding method based on PSI5 interface, characterized in that: include: S1: Collect historical signal data and analog signals output by sensors under different working environment conditions; S2: Extract features of the analog signal to obtain feature parameters; Preprocessing the historical signal data to obtain predicted data; S3: Dynamically calibrating the characteristic parameters using a calibration rule to obtain an accurate analog signal; S4: generating a decoding rule adapted to the current signal using a decision tree algorithm according to the precise analog signal and the PSI5 protocol standard; S5: Decoding the precise analog signal according to the decoding rule to obtain valid data; S6: Compare the valid data with the expected data, adjust the calibration rules and decision tree algorithm parameters according to the deviation, and obtain accurate valid data.

2. The method of dynamic calibration and adaptive signal decoding based on PSI5 interface according to claim 1, characterized in that: In step S1, the analog signal is converted into a digital signal through a PSI5 interface. The digital signal includes a plurality of data frames, and each of the data frames includes a synchronization bit, a data bit, and a check bit.

3. The method of dynamic calibration and adaptive signal decoding based on PSI5 interface according to claim 1, characterized in that: In step S2, the characteristic parameters include the initial amplitude, frequency, phase and data frame format of the analog signal; the preprocessing includes: performing outlier processing, One-Hot encoding, time encoding, data normalization, and adding Gaussian noise processing on the historical signal data.

4. The method of dynamic calibration and adaptive signal decoding based on PSI5 interface according to claim 1, characterized in that: In step S3, dynamically calibrating the characteristic parameters using calibration rules includes: S31: Set the structural parameters of the time series integration model and collect real-time environment parameters; S32: Using the predicted data to train a time series integration model; S33: Inputting the real-time environmental parameters into the trained time series integration model to predict current calibration parameters, converting the current calibration parameters into operable calibration rules, and dynamically adjusting the signal characteristic parameters according to the calibration rules.

5. The method of dynamic calibration and adaptive signal decoding based on PSI5 interface according to claim 4, characterized in that: In step S32, training the time series integration model includes: S321: Divide the predicted data into a training set 1 and a test set 1; S322: setting hyperparameters of the current time series integration model according to the structural characteristics of the time series integration model and characteristic parameters of the predicted data; S323: Convert the training set into a low-dimensional learnable space using an embedder of a temporal ensemble model; S324: In the low-dimensional learnable space, the discriminator updates the hyperparameters of the current time series integration model according to the accuracy of discriminating whether the received data is from the training set 1 or the fake data generated by the generator using the gradient descent algorithm until the comprehensive evaluation index of the time series integration model decreases for two consecutive rounds and the hyperparameter update is stopped; S325: Use a test set to test the time series integration model until the trained time series integration model meets a preset threshold, and use the generator to generate new data and merge it with the original predicted data to obtain new predicted data.

6. The method of dynamic calibration and adaptive signal decoding based on PSI5 interface according to claim 5, characterized in that: In step S324, training the time series integration model includes: during the hyperparameter update process, the generator continuously generates false data in an attempt to deceive the discriminator, and the discriminator finds the false data output by the generator, and the two repeatedly perform adversarial training. After each round of adversarial training, the time series integration model is tested using a test set, and the model performance indicators and changes in hyperparameters are monitored until the time series integration model stops updating the hyperparameters.

7. The method of dynamic calibration and adaptive signal decoding based on PSI5 interface according to claim 1, characterized in that: In step S4, the specific steps of using the decision tree algorithm to generate a decoding rule adapted to the current signal are: S41: Setting parameters of the decision tree algorithm according to the number of the characteristic parameters; S42: constructing a training set 2 and a test set 2 by combining the characteristic parameters of the analog signal and the correct decoding under the corresponding preset decoding rules, and inputting the training set 2 into the decision tree algorithm; S43: Splitting the best feature into nodes and constructing branches to form a decision tree. The categories corresponding to the leaf nodes of the decision tree are preliminary decoding rules. S44: Using a set of current analog signals with known correct decoding results, the current analog signals are decoded according to preliminary decoding rules to obtain current analog signal decoding rules.

8. The method of dynamic calibration and adaptive signal decoding based on PSI5 interface according to claim 1, characterized in that: In step S5, decoding the precise analog signal includes: S51: Finding the starting frame in the precise analog signal according to the definition of the signal frame structure in the decoding rule; S52: From the position of the starting frame, check whether the precise analog signal frame after the starting frame is complete according to the decoding rule, and if so, keep it unchanged; otherwise, repair the precise analog signal frame using a signal repair algorithm; S53: extracting analog signal data corresponding to the data segment from the complete accurate analog signal frame according to the position and length range of the data segment in the signal frame in the decoding rule; S54: converting the data segment of the analog signal into digital signal data according to the decoding conversion method for the data segment in the decoding rule; S55: parsing and restoring the digital signal data according to the data format in the decoding rules to obtain valid data.

9. The method of dynamic calibration and adaptive signal decoding based on PSI5 interface according to claim 8, characterized in that: In step S53, the data segment includes: Position-related data: used to indicate the specific position of the data segment in the entire signal frame; Length range data: the length range occupied by the data segment itself; Coding characteristic data: the coding characteristics of the data segment in the form of analog signals; Decoding conversion data: describes the specific conversion method of how the data segment is converted from analog signal to digital signal; Data format data: The data format requirements followed when the data segment is finally parsed and restored.

10. The method of dynamic calibration and adaptive signal decoding based on PSI5 interface according to claim 1, characterized in that: In step S6, after adjustment, the parsed parameters are error checked to determine whether the error check result exceeds a threshold. If so, the redundant data is used to compensate for the valid data.

Images